If they are, they aren't producing anything useful with it. Just look around - do you see a sudden increase in actually useful software alongside the AI boom?
What they are mostly doing is a snake-eating-it-own-tail million lines of code LLM harness to burn tokens faster to write more code... to write a 10 million lines of code LLM harness. Or endlessly bikeshedding the perfect LLM-powered bespoke personal knowledge base.
In normal software engineering jobs, we're debugging problems a bit quicker, we're writing boilerplate faster, we have a lot of questionable new test suites... but the game is more or less the same as it was before
Putting it a different way, it won't matter if the firms who went too deep at the very beginning are fucked if the rational are forced to succumb to the market pressures created by the irrational and thus are reluctantly pushed to adopt AI-first workflows for appearance's sake in order to survive anyway. Because then everybody will be likewise fucked and completely dependent on AI, despite it being a subpar development paradigm with respect to robustness of the systems under development. History has taught us that it is adoption dynamics not capability that determines the winning paradigm or technology (Betamax vs VHS is one historical example. Javascript vs everything else is another one).
(We know it's a subpar development paradigm with respect to robustness because the entire coding agent paradigm turns the most knowledgeable programmer into a person "who doesn't know what they don't know" because development speed far exceeds their ability to reason about the codebase and the underlying SOTA models that they depend on to fix the bugs that the model itself has introduced are at best unreliable narrators with no objective evidence of correctness or deterministic behavior.)
It may be three years or so before the new compute-in-memory devices fully make it out of the lab and increase efficiency by about 100 times, allowing us to deploy models with human level complexity (100T vs current 10T SOTA) at scale.
That is when I realize I'm spending too much time on HN. Because it is really only here that this vibe is so strong. My impression is that there is a lot of motivated reasoning in the folks that frequent HN.
It may also be that I work a boring job. If I turned up our code output by 10x it would not improve anything about our product. People who are pumping out dramatically more code have to be in an entirely different world. Or, you know, they're full of shit.